• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

big hit on points.

D

dannifood4less

Limp Gawd
Joined
Sep 4, 2008
Messages
132
so a few days ago all of my clients broke at the same time. (my gpu and my dual SMP setup were EUEing) i got everything back up and running but later that night the gpu couldnt get any work.... and now were on day 3... ive seen the GPU servers go down before, but COME ON,, 3 days and i cant get any work for my GPU?!

[14:17:11] + Could not connect to Work Server
[14:17:11] - Attempt #15 to get work failed, and no other work to do.
Waiting before retry.

is this happening to everyone or just some people? i know were all tired of seeing threads complaining about the GPU servers, but i have never seen my GPU looking for work for longer than a day..

 
yeah ive been getting it all night.. it periodicly works.. been happening since about 4 am PST.. atleast thats when i finally noticed it.. usually connects after then 15th attempt..

was just about to create a thread for this.. so thanks for doing it for me.. :p

and btw its both servers..

CPU client..
[14:46:37] + Attempting to get work packet
[14:46:37] - Connecting to assignment server
[14:46:37] - Successful: assigned to (0.0.0.0).
[14:46:37] + News From Folding@Home: Welcome to Folding@Home
[14:46:37] Work Unit has an invalid address.
[14:46:37] - Error: Attempt #9 to get work failed, and no other work to do.
Waiting before retry.

not sure why the times wrong on the logs.. but meh..
 
maybe the problems gone now.. but my gpu client just sent/received data first try.. last 5 WU's have been the 5506 wu though..looks like they are really pushing that project for the last 12 hours..

cpu servers still down.. it keeps assigning me with ip 0.0.0.0
 
ALL of my GPUs stopped for some time last night. But for some damn reason my mom's GPU stayed dormant ALL night, and I just had to restart it to get a block. Same problem as you guys. Can't get work waiting for retry #15 I think. It burns me up that one of my GPUs sat dormant ALL F*#%)#% NIGHT. That's a nice chunk of my PPD today!!!
 
well just think everyone pretty much suffered the same as you mrboomy.. luckly i got a few WU's in during the night but after 4:30 am its been hit and miss..


update: looks like the cpu clients working again.. just got a 4420 wu only 150k step wu though.. and amber core specific.. i think this might be one of those 2xpoints Wu's that i read on their site..

update 2: gpu servers still hit and miss.. i did get a WU but still getting this error...

[16:34:21] + Attempting to send results [October 27 16:34:21 UTC]
[16:34:21] - Couldn't send HTTP request to server
[16:34:21] (Got status 503)
[16:34:21] + Could not connect to Work Server (results)
[16:34:21] (171.67.108.25:8080)
[16:34:21] + Retrying using alternative port
[16:34:21] - Couldn't send HTTP request to server
[16:34:21] (Got status 503)
[16:34:21] + Could not connect to Work Server (results)
[16:34:21] (171.67.108.25:80)
[16:34:21] Could not transmit unit 09 to Collection server; keeping in queue.
[16:34:25] Will resume from checkpoint file
[16:34:25] Working on p5506_supervillin_e1
[16:34:26] Client config found, loading data.
[16:34:26] Resuming from checkpoint
 
It would appear that everyone is in the same boat. Points are down across the board. As a team, we only did 2.6 Million points yesterday. :eek:


2.png
 
It was a weekend so servers get backed up :(

At least it is consistent

 
This reminds me of a drunk walking. Two steps forward and one step back. Certainly makes the journey take longer.
 
arg.. still cant send data back.. now on my 4th GPU WU.. with 3 waiting in queue to be sent..
 
ya i diddnt want to post about it at first cuz i knew it always gets like this on weekends, but i thought it was a bit more than usual. so i wanted to see if it was everyone or just my gpu. i got a WU now lets see what happens when it finishes it. :D


 
Darn servers again. Maybe Stanford should use all the 'extra resources' recently acquired to shore up their infrastructure instead of contributing to the flood. :rolleyes:

 
APOLLO said:
Darn servers again. Maybe Stanford should use all the 'extra resources' recently acquired to shore up their infrastructure instead of contributing to the flood. :rolleyes:
Click to expand...

LOL ... Those ''extra resources" is a computer cluster not servers.
They only have it for a very short time.

You also have to remember that since the nVidia cleint is having more EUE's right now, it is causing increased server load.

http://fah-web.stanford.edu/serverstat.html

If you remember back in July, 6 ATI users were able to bring down a server with EUEing cleints. http://foldingforum.org/viewtopic.php?f=24&t=4242

Now, think about about all those GPU2 clients now ....

At the time of this post, the ATI server load is at 22.73 which is really really high.

Once the EUE's stop or lessen, the server load will reduce.

ATI just got our second GPU server and I am sure nVidia will get another one when time / funds permit.

 
EvilAlchemist said:
LOL ... Those ''extra resources" is a computer cluster not servers. They only have it for a very short time.
Click to expand...
Yes, but the problem is affecting everyone at present. Murphy's Law seems to be operating on overtime lately. Chances are when the cluster will be gone most of the problems the project is facing now will also have diminished. Such is the nature of fortune.

You also have to remember that since the nVidia cleint is having more EUE's right now, it is causing increased server load.
Click to expand...
Yes, I'm well aware of it. I started a thread about that very topic last week. The point is that it hasn't been fixed and now an additional load of 200,000 PPD was introduced from a single source recently that isn't helping the current server load situation. As I mentioned elsewhere, I don't think it's negative for such a generous donation to appear serendipitously like it has, only poor timing with everything else transpiring at the same time.

If you remember back in July, 6 ATI users were able to bring down a server with EUEing cleints. http://foldingforum.org/viewtopic.php?f=24&t=4242

Now, think about about all those GPU2 clients now ....
Click to expand...
Actually, I don't recall or know of that case, but it only underscores the precarious condition of Stanford's server infrastructure more than anything else, to tell you the truth.

At the time of this post, the ATI server load is at 22.73 which is really really high.
Click to expand...
I don't ordinarily experience delays with my Ati client receiving work. The only issue I have with my Ati client is the occasional backlogged WUs which don't get dispatched to Stanford despite the fact most newer WUs upload fine. The WUs in queue for some reason never upload and eventually expire.

My main gripe is with the nVidia client but at least it doesn't have the expired backed-up WU syndrome. I was contemplating adding a second Ati card a few months down the road, but I have changed my mind since I saw what was happening with this one. In fact, I've put off all GPU purchases for the indefinite future.

Once the EUE's stop or lessen, the server load will reduce.
Click to expand...
That's exactly my point. Stanford releases a bad core, EUEs skyrocket but there isn't the funds to acquire more hardware to handle the situation exacerbating the problem of reduced production when clients mass idle. Yet, Stanford doesn't revert to a tried-and true older core. Approximately, 20% of the nVidia-targeted work I receive concludes in an EUE. On a bad day I've seen more than half of my GPU WUs EUE. I'm aware that the newer WUs require the problematic core, but why not delay issuing those WUs until everything is determined working sound?

ATI just got our second GPU server and I am sure nVidia will get another one when time / funds permit.
Click to expand...
I see. How many servers are dedicate for each GPU client, BTW??
 
nVidia has 4 servers, ATI has 2.

Which servers run what is Here. but to quote the post .............
Here is a summary of the servers used, and which client will used them.

Assignment server (All CPU and SMP clients on port 8080)
171.65.103.93
Shown as assign.stanford.edu:8080 in the clients.

Secondary assignment server (All CPU and SMP clients on port 80)
171.64.65.121
Shown as assign2.stanford.edu:80 in the clients.

Assignment server for GPU clients
171.65.103.98
Shown as assign-GPU.stanford.edu:8080 in the clients.

Assignment server for PS3 clients
171.65.103.96

Secondary assignment server for PS3 clients
171.65.103.102

Collection servers
171.65.103.100
171.64.122.76
171.64.122.86
Each server can be used by all clients. See CS and CSlisted columns in the Server Status page to know which Collection Server each work server will use.

SMP work servers
171.64.65.56 (mostly used to serve Linux SMP)
171.64.65.63
171.64.65.64 (mostly used to server Windows SMP)

ATI GPU2 work servers
171.64.65.103
171.64.65.102

nVidia GPU2 work servers
171.64.65.20
171.64.122.74
171.64.65.106
171.67.108.11

Every server not listed above will be used by the uniprocessor client.

/!\ This list is subject to changes /!\ ... please notify me if a server is missing or has been moved to another task.

Luck ................. :D
 
APOLLO, why don't you go sign up for the Beta Testing on the Folding Forums ?

I know venting here feels good, but why don't you post it over there where they can see it and respond.

If you get into beta, then you will really have their ear and can express feedback on the client , cores, servers, and all of that ...

Then you will have a good understand of what is going on ... how much work goes into a core & client .. what testing is like .....and what is going wrong / right and what is being done ...

I am in there, and Xilikon , & and i think Sunin too
 
EvilAlchemist said:
APOLLO, why don't you go sign up for the Beta Testing on the Folding Forums ?

I know venting here feels good, but why don't you post it over there where they can see it and respond.

If you get into beta, then you will really have their ear and can express feedback on the client , cores, servers, and all of that ...

Then you will have a good understand of what is going on ... how much work goes into a core & client .. what testing is like .....and what is going wrong / right and what is being done ...

I am in there, and Xilikon , & and i think Sunin too
Click to expand...
Evil, I respect you Sunin, Xilikon very much and I did sign up sort of, but 7im permanently barred my involvement with the beta team back in the summer.
 
i havent had an eue error at all.. but god this constant waiting is driving me insane..
 
You must log in or register to reply here.
Back
Top